cement stabilization of embankment materials final report

Cement Stabilization of Embankment Materials

Final Report

JULY 2015

Sponsored byIowa Department of Transportation

About the Center for Earthworks Engineering Research

The mission of the Center for Earthworks Engineering Research (CEER) at Iowa State University is to be the nation’s premier institution for developing fundamental knowledge of earth mechanics, and creating innovative technologies, sensors, and systems to enable rapid, high quality, environmentally friendly, and economical construction of roadways, aviation runways, railroad embankments, dams, structural foundations, fortifications constructed from earth materials, and related geotechnical applications.

Disclaimer Notice

The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. The opinions, findings and conclusions expressed in this publication are those of the authors and not necessarily those of the sponsors.

The sponsors assume no liability for the contents or use of the information contained in this document. This report does not constitute a standard, specification, or regulation.

The sponsors do not endorse products or manufacturers. Trademarks or manufacturers’ names appear in this report only because they are considered essential to the objective of the document.

Non-Discrimination Statement

Iowa State University does not discriminate on the basis of race, color, age, ethnicity, religion, national origin, pregnancy, sexual orientation, gender identity, genetic information, sex, marital status, disability, or status as a U.S. veteran. Inquiries regarding non-discrimination policies may be directed to Office of Equal Opportunity, Title IX/ADA Coordinator, and Affirmative Action Officer, 3350 Beardshear Hall, Ames, Iowa 50011, 515-294-7612, email [email protected].

Iowa Department of Transportation Statements

Federal and state laws prohibit employment and/or public accommodation discrimination on the basis of age, color, creed, disability, gender identity, national origin, pregnancy, race, religion, sex, sexual orientation or veteran’s status. If you believe you have been discriminated against, please contact the Iowa Civil Rights Commission at 800-457-4416 or the Iowa Department of Transportation affirmative action officer. If you need accommodations because of a disability to access the Iowa Department of Transportation’s services, contact the agency’s affirmative action officer at 800-262-0003.

The preparation of this report was financed in part through funds provided by the Iowa Department of Transportation through its “Second Revised Agreement for the Management of Research Conducted by Iowa State University for the Iowa Department of Transportation” and its amendments.

The opinions, findings, and conclusions expressed in this publication are those of the authors and not necessarily those of the Iowa Department of Transportation.

Technical Report Documentation Page

1. Report No. 2. Government Accession No. 3. Recipient’s Catalog No.

4. Title and Subtitle 5. Report Date

Cement Stabilization of Embankment Materials July 2015

6. Performing Organization Code

7. Author(s) 8. Performing Organization Report No.

Shengting Li, David White, and Pavana Vennapusa

9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)

Center for Earthworks Engineering Research

Iowa State University

2711 South Loop Drive, Suite 4600

Ames, IA 50010-8664

11. Contract or Grant No.

12. Sponsoring Organization Name and Address 13. Type of Report and Period Covered

Iowa Department of Transportation

800 Lincoln Way

Ames, IA 50010

Final Report

14. Sponsoring Agency Code

15. Supplementary Notes

Visit www.ceer.iastate.edu for color PDFs of this and other research reports.

16. Abstract

Embankment subgrade soils in Iowa are generally rated as fair to poor as construction materials. These soils can exhibit low bearing

strength, high volumetric instability, and freeze/thaw or wet/dry durability problems. Cement stabilization offers opportunities to

improve these soils conditions. The objective of this study was to develop relationships between soil index properties, unconfined

compressive strength and cement content. To achieve this objective, a laboratory study was conducted on 28 granular and non-granular

materials obtained from 9 active construction sites in Iowa. The materials consisted of glacial till, loess, and alluvium sand. Type I/II

portland cement was used for stabilization. Stabilized and unstabilized specimens were prepared using Iowa State University 2 in. by

2 in. compaction apparatus. Specimens were prepared, cured, and tested for unconfined compressive strength (UCS) with and without

vacuum saturation. Percent fines content (F200), AASHTO group index (GI), and Atterberg limits were tested before and after

stabilization. The results were analyzed using multi-variate statistical analysis to assess influence of the various soil index properties on

post-stabilization material properties.

Results indicated that F200, liquid limit, plasticity index, and GI of the materials generally decreased with increasing cement content. The

UCS of the stabilized specimens increased with increasing cement content, as expected. The average saturated UCS of the unstabilized

materials varied between 0 and 57 psi. The average saturated UCS of stabilized materials varied between 44 and 287 psi at 4% cement

content, 108 and 528 psi at t 8% cement content, and 162 and 709 psi at 12% cement content. The UCS of the vacuum saturated

specimens was on average 1.5 times lower than that of the unsaturated specimens. Multi-variate statistical regression models are

provided in this report to predict F200, plasticity index, GI, and UCS after treatment, as a function of cement content and soil index

properties.

17. Key Words 18. Distribution Statement

cement stabilization—cohesive soils—embankment soils—mix design No restrictions.

19. Security Classification (of this

report)

20. Security Classification (of this

page)

21. No. of Pages 22. Price

Unclassified. Unclassified. 64 NA

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

CEMENT STABILIZATION OF EMBANKMENT

MATERIALS

Final Report

July 2015

Principal Investigator

David J. White, Ph.D., P.E.

Richard L. Handy Professor of Civil Engineering

Director, Center for Earthworks Engineering Research

Co-Principal Investigator

Pavana K. R. Vennapusa, Ph.D., P.E.

Assistant Director, Center for Earthworks Engineering Research

Research Assistant

Shengting Li

Authors

Shengting Li, David White, and Pavana Vennapusa

Sponsored by

the Iowa Department of Transportation

Preparation of this report was financed in part

through funds provided by the Iowa Department of Transportation

through its Research Management Agreement with the

Institute for Transportation

A report from

Center for Earthworks Engineering Research

Iowa State University

2711 South Loop Drive, Suite 4600

Ames, IA 50010-8664

Phone: 515-294-7910

Fax: 515-294-0467

www.ceer.iastate.edu

v

TABLE OF CONTENTS

ACKNOWLEDGMENTS ............................................................................................................ IX

EXECUTIVE SUMMARY .......................................................................................................... XI

CHAPTER 1: INTRODUCTION ....................................................................................................1

CHAPTER 2: TESTING METHODS .............................................................................................2

Soil Index Properties ............................................................................................................2 Proctor Compaction .............................................................................................................3 ISU 2 in. by 2 in. Compaction .............................................................................................4 Unconfined compressive strength (UCS) ............................................................................5

Statistical Analysis ...............................................................................................................7

CHAPTER 3: MATERIALS ...........................................................................................................8

CHAPTER 4: RESULTS AND DISCUSSION .............................................................................20

Fines Content (F200) ...........................................................................................................23 Atterberg Limits .................................................................................................................25 AASHTO Group Index ......................................................................................................32

Unconfined Compressive Strength ....................................................................................35

CHAPTER 5: SUMMARY OF KEY FINDINGS.........................................................................41

REFERENCES ..............................................................................................................................43

APPENDIX A: LABORATORY TEST RESULTS......................................................................45

APPENDIX B: IOWA DOT PROPOSED INSTRUCTIONAL MEMORANDUM FOR

CEMENT STABILIZATION OF SOILS ..........................................................................49

vi

LIST OF FIGURES

Figure 1. Theoretical relationship between water content and dry density (Reproduced from

Li and Sego 2000) ................................................................................................................3

Figure 2. ISU 2 in. by 2 in. specimen compaction ...........................................................................4

Figure 3. Specimen failure after measurement of UCS ...................................................................6

Figure 4. Vacuum saturation of cement stabilized specimens .........................................................7

Figure 5. Grain size distribution of embankment materials obtained from Polk County ..............15

Figure 6. Grain size distribution of embankment materials obtained from Warren County .........15

Figure 7. Grain size distribution of embankment materials obtained from Linn County 77 .........16

Figure 8. Grain size distribution of embankment materials obtained from Linn County 79 .........16

Figure 9. Grain size distribution of embankment materials obtained from Mills County .............17

Figure 10. Grain size distribution of embankment materials obtained from Pottawattamie

County ................................................................................................................................17

Figure 11. Grain size distribution of embankment materials obtained from Woodbury

County I-29 ........................................................................................................................18

Figure 12. Grain size distribution of embankment materials obtained from Scott County ...........18


County US 20 .....................................................................................................................19

Figure 14. F200 versus cement content ...........................................................................................23

Figure 15. F200 versus cement content (continued) ........................................................................24

Figure 16. Comparison of measured F200 and predicted F200 .........................................................25

Figure 17. Plasticity chart with results of unstabilized soils ..........................................................26

Figure 18. Plasticity chart with results of 4% cement stabilized soils ...........................................26

Figure 19. Plasticity chart with results of 8% cement stabilized soils ...........................................27

Figure 20. Plasticity chart with results of 12% cement stabilized soils .........................................27

Figure 21. PI versus F200 for unstabilized and stabilized soils .....................................................28

Figure 22. LL and PI versus cement content .................................................................................29

Figure 23. LL and PI versus cement content (continued) ..............................................................30

Figure 24. LL and PI versus cement content (continued-2) ...........................................................31

Figure 25. Comparison of measured PI and predicted PI ..............................................................32

Figure 26. AASHTO group index versus cement content .............................................................33

Figure 27. AASHTO group index versus cement content (continued) ..........................................34

Figure 28. Comparison of measured group index and predicted group index ...............................35

Figure 29. Unsaturated and vacuum saturated UCS versus cement content ..................................36

Figure 30. Unsaturated and vacuum saturated UCS versus cement content (continued) ..............37

Figure 31. Unsaturated and vacuum saturated UCS versus cement content (continued-2) ...........38

Figure 32. Comparison of measured unsaturated UCS and predicted unsaturated UCS ...............40

Figure 33. Comparison of measured vacuum saturated UCS and predicted UCS.........................40

vii

LIST OF TABLES

Table 1. Testing methods used in the study .....................................................................................2

Table 2. Number of drop-hammer blows (O’Flaherty et al. 1963) ..................................................5

Table 3. Soil index properties of embankment materials obtained from Polk County ....................9

Table 4. Soil index properties of embankment materials obtained from Warren County and

Linn County 79 ..................................................................................................................10

Table 5. Soil index properties of embankment materials obtained from Linn County 77.............11

Table 6. Soil index properties of embankment materials obtained from Pottawattamie

County and Woodbury County I-29 ..................................................................................12

Table 7. Soil index properties of embankment materials obtained from Scott County and

Mills County ......................................................................................................................13

Table 8. Soil index properties of embankment materials obtained from Woodbury County

US 20 .................................................................................................................................14

Table 9. Summary of soil index properties and Iowa DOT suitability classifications at

different cement contents ...................................................................................................20

Table 10. Multi-variate analysis results to predict F200 after cement stabilization ........................24

Table 11. Multi-variate analysis results to predict PI after cement stabilization ...........................31

Table 12. Multi-variate analysis results to predict GI after cement stabilization ..........................34

Table 13. Multi-variate analysis results to predict unsaturated UCS.............................................39

Table 14. Multi-variate analysis results to predict vacuum saturated UCS ...................................39

ix

ACKNOWLEDGMENTS

This project was sponsored by the Iowa Department of Transportation (DOT). Melissa Serio and

Steve Megivern of the Iowa DOT provided valuable feedback through the course of this project.

Chao Chen of Iowa State University assisted in performing laboratory testing. All of their efforts

are greatly appreciated.

xi

EXECUTIVE SUMMARY

Embankment subgrade soils in Iowa are generally rated as fair to poor as construction materials.

These soils can exhibit low bearing strength, high volumetric instability, and freeze/thaw or

wet/dry durability problems. Cement stabilization offers opportunities to improve these soils

conditions. The Iowa Department of Transportation (DOT) is considering the use of portland

cement as an additive for stabilizing embankment materials in situ. The objective of this study

was to develop relationships between soil index properties, unconfined compressive strength and

cement content. To achieve this objective, a laboratory study was conducted on 28 granular and

non-granular materials obtained from 9 active construction sites in Iowa. The materials consisted

of glacial till, western Iowa loess, and alluvium sand. Type I/II portland cement was used in this

study.

Stabilized and unstabilized specimens were prepared using Iowa State University 2 in. by 2 in.

compaction apparatus. Specimens were prepared, cured, and tested for unconfined compressive

strength (UCS) with and without vacuum saturation. Percent fines content (F200), AASHTO

group index (GI), and Atterberg limits were tested before and after stabilization. The results were

analyzed using multi-variate statistical analysis to assess influence of the various soil index

properties on post-stabilization material properties. Following the results of this study, a draft

laboratory testing and evaluation procedure is provided in the Appendix of this report.

Key findings from the test results and analysis are as follows:

F200 of the material decreased with increasing cement content for a majority of the soils.

The percent cement content, F200 before treatment, and liquid limit were found to be

statistically significant in predicting the F200 after treatment. The multi-variate model

showed an R2 of about 0.9 and RMSE of about 7% in predicting the F200 after treatment.

With the exception of a few materials, the liquid limit and plasticity index of all materials

decreased with increasing cement content. The one untreated soil classified as

“unsuitable”, classified as “suitable” after stabilized with cement. Some of the untreated

soils that were classified as “select”, classified as “suitable” after stabilized with cement.

The classifications changed because of reduction in plasticity index. All soils classified as

“suitable” at 12% cement content because of no plasticity. The percent cement content

and clay content parameters were found to be statistically significant in predicting the

plasticity index of materials after stabilization. The multi-variate model showed an R2 of

about 0.5 and RMSE of about 5%.

The GI values decreased with increasing cement content for a majority of the soils. The

percent cement content, F200, liquid limit, and plasticity index parameters were found to

be statistically significant in predicting the group index values after treatment. The multi-

variate model showed an R2 of about 0.7 and RMSE of about 3.

xii

The UCS of specimens increased with increasing cement content, as expected. The

average saturated UCS of the unstabilized materials varied between 0 and 57 psi. The

average saturated UCS of stabilized materials varied between 44 and 287 psi at 4%

cement content, 108 and 528 psi at t 8% cement content, and 162 and 709 psi at 12%

cement content. The draft laboratory testing and evaluation procedure for cement

stabilization mix design provided in Appendix B targets a 100 psi saturated unconfined

compressive strength. The UCS of the saturated specimens was on average 1.5 times

lower than of the unsaturated specimens.

The percent cement content, sand content, fines content, and liquid limit were found to be

statistically significant in predicting unsaturated and vacuum saturated UCS. The models

showed an R2 of about 0.85 and RMSE of about 75 psi for vacuum saturated specimens

and 97 psi for unsaturated specimens.

1

CHAPTER 1: INTRODUCTION

Embankment subgrade soils in Iowa are generally rated as fair to poor as construction materials,

with a majority of the soils classifying as A-4 to A-7-6 according to the AASHTO (2004). These

soils can exhibit low bearing strength, high volumetric instability, and freeze/thaw or wet/dry

durability problems. Cement stabilization offers opportunities to improve these soils conditions.

The Iowa Department of Transportation (DOT) is considering the use of portland cement as an

additive for stabilizing embankment materials in situ.

The objective of this study was to develop relationships between soil index properties,

unconfined compressive strength and cement content. To achieve this objective, a laboratory

study was conducted on 28 granular and non-granular materials obtained from 9 active

construction sites in Iowa. The materials consisted of glacial till, western Iowa loess, and

alluvium sand. Type I/II portland cement was used in this study.

Stabilized and unstabilized specimens were prepared using Iowa State University 2 in. by 2 in.

compaction apparatus, and were tested for unconfined compressive strength (UCS) with and

without vacuum saturation. Soil index properties such as the percent fines passing the No. 200,

AASHTO group index, and Atterberg limits were assessed before and after stabilization. The

results were analyzed using multi-variate statistical analysis to assess influence of the various

soil index properties on post-stabilization material properties. Following the results of this study,

a draft laboratory testing and evaluation procedure is provided in the Appendix of this report.

2

CHAPTER 2: TESTING METHODS

The ISU research team performed a series of laboratory tests on materials collected from various

embankment construction sites in Iowa. Table 1 summarizes the test standards and procedures

used in this study.

Table 1. Testing methods used in the study

Test standard or reference Description of test

ASTM C593-06 Vacuum saturation

ASTM D422-63 Particle-size analysis

ASTM D698-12 Standard Proctor

ASTM D854-14 Specific gravity

ASTM D1140-14 Fines content passing the No. 200 sieve

ASTM D1557-12 Modified Proctor

ASTM D1633-00 Unconfined compressive strength (UCS)

ASTM D2487-11 Unified soil classification system (USCS)

ASTM D3282-09

American Association for State Highway

and Transportation Officials (AASHTO)

classification system

ASTM D4318-10 Atterberg limits

O’Flaherty et al. (1963) ISU 2 in. x 2 in. compaction

Soil Index Properties

Particle-size analysis tests were conducted in accordance with ASTM D422-63 (ASTM 2014a).

The distribution of particle sizes larger than 75 µm (opening size of the No. 200 sieve) is

determined by sieving, and the distribution of particle sizes smaller than 75 µm is determined by

the hydrometer method.

Percent passing the No. 200 sieve (F200) was determined in accordance with ASTM D1140-14

(ASTM 2014e).

Atterberg limits tests to determine liquid limit (LL), plastic limit (PL), and plasticity index (PI)

were conducted in accordance with ASTM D4318-10 (ASTM 2014b), using the “wet

preparation” method. LL tests were performed using the multipoint method.

Utilizing the results of the above testing, each sample was classified using the Unified Soil

Classification System (USCS) in accordance with ASTM D2487-11 (ASTM 2011b) and the

3

AASHTO classification system in accordance with ASTM D3282-09 (ASTM 2009). AASHTO

group index (GI) values were determined in accordance with AASHTO (2004) using Eq. 1:

GI = (F200 – 35) x [0.2 + 0.005 x (LL-40)] + 0.01 x (F200 -15) x (PI – 10) (1)

Specific gravity tests were conducted in accordance with ASTM 854-14 Method A – procedure

for moist specimens (ASTM 2014).

Proctor Compaction

The relationship between moisture and dry unit weight of embankment materials was determined

in accordance with ASTM D698-12 (ASTM 2012a) and ASTM D1557-12 (ASTM 2012b). The

appropriate method was chosen based upon the grain-size distributions for each material.

Method A was applicable for all soil materials. The tests were performed at five different

moisture contents and the optimum moisture-density characteristics were obtained based upon

the Li and Sego Fit model (Li and Sego 2000) fit to the data. Figure 1 shows the model, and

Eq. 2 shows the relationship and the relevant parameters.

Figure 1. Theoretical relationship between water content and dry density (Reproduced

from Li and Sego 2000)

𝛾𝑑 (w) = 𝐺𝑠 𝛾𝑤

(1+𝑤 𝐺𝑠

𝑆𝑚−𝑆𝑚 (𝑤𝑚−𝑤𝑤𝑚

)𝑛+1 (𝑤𝑚𝑛 +𝑝𝑛

(𝑤𝑚−𝑤)+𝑝𝑛)

) (2)

Water content (%)

0 10 20 30 40

Dry

density (

pcf)

80

85

90

95

100

105

110

115

ZAVC (S=100%)

S=Sm

CompactionCurve

Approach line

max

dd

wopt

wm

4

where,

γd = dry density of the soil;

Gs = specific gravity of the soil;

γw = density of water;

w = moisture content of the soil;

wm = moisture content at Sm;

Sm = maximum of saturation; and

n and p = shape factors

Sm defines the maximum saturation boundary of the model on the wet side of optimum and the

corresponding moisture content is defined as wm. It can be determined from the wet side of the

compaction curve running parallel to the zero air void curve. The parameter n is referred to as the

shape factor, which affects the dome portion of the compaction curve. When n is increased, the

dome portion of the compaction curve becomes sharper; and when n is decreased, the dome

portion tends to become flatter. The parameter p influences the width of the upper portion of the

curve. In determining the best fit parameters, Sm and wm were first determined based on the data

to establish the boundary of the curve, and shape factor n and p were adjusted until a maximum

correlation coefficient (R2) between the measured and the predicted values is achieved.

ISU 2 in. by 2 in. Compaction

ISU 2 in. by 2 in. compaction apparatus is described in O’Flaherty et al. (1963). The test

procedure was used to prepare 2 in. diameter by 2 in. height (2 x 2) samples for UCS testing

(Figure 2).

Figure 2. ISU 2 in. by 2 in. specimen compaction

5

Samples were compacted at their respective standard Proctor optimum moisture content. For

cement treated materials, the optimum moisture content was determined using Eq. 3 with a water

to cement (w/c) ratio of 0.25:

wopt soil + cement = [(% cement added by weight) × (w/c ratio)] + wopt soil (3)

The test procedure involved placing loose material in the compaction apparatus and dropping a

5 lb. hammer from a drop height of about 12 in. in a 2 in. diameter steel mold. O’Flaherty et al.

(1965) provided guidance on the number of blows required to obtain standard Proctor densities

for different soil types, as summarized in Table 2. The number of blows were selected based on

the soil type and equal number of blows were applied on both sides of the sample, to compact the

sample uniformly.

Table 2. Number of drop-hammer blows (O’Flaherty et al. 1963)

AASHTO Soil Type Total number of

drop-hammer blows

A-7 and A-6 6

A-4 7

A-3, A-2, and A-1 14

After compaction, the 2 x 2 specimens were sealed using a saran wrap and aluminum foil, and

were placed in a Ziploc bag. Cement stabilized specimens were cured for 7 days at 110oF, to

simulate 28 day curing strength (Winterkkorn and Pamukcu 1990). Unstabilized specimens were

tested shortly after compaction (no curing). Three samples were prepared at each cement content.

Unconfined compressive strength (UCS)

The cured specimens were tested for UCS (Figure 3) in general accordance with ASTM D 1633-

00 (ASTM 2007). The standard requires use of either 4 in. diameter by 4.584 in. height Proctor

samples with a height to diameter (h/d) ratio of 1.15 or 2.8 in. diameter by 5.6 in. height samples

with an h/d ratio of 2.0. Instead, 2 x 2 specimens were used in this study which have an h/d ratio

of 1.0. This decision was made based on a previous study (White et al. 2005) with design of fly

ash soil mixtures, where it was concluded that UCS can be determined using any of these three

sample sizes: h/d = 1.0 (2x2 sample), 1.15 (Proctor), and 2.0 (2.8 in. x 5.6 in.).

Based on laboratory testing on fly ash soil mixtures, White et al. (2005) concluded that the UCS

determined from 2 x 2 specimens can be multiplied by 0.86 to correlate with UCS of Proctor

sized samples (h/d = 1.15) or 0.90 to correlate with samples that have h/d = 2. ASTM D1633-00

also provides a similar guidance in relating UCS on samples with h/d=1.15 to samples with

h/d=2 as follows: “If desired, make allowance for the ratio of height to diameter (h/d) by

multiplying the compressive strength of Method B specimens [with h/d = 2.0] by factor 1.10. This

converts the strength for an h/d ratio of 2.00 to that for the h/d ratio of 1.15 commonly used in

routine testing of soil-cement.” The ASTM D1633 correction factor provides a higher UCS value

6

for samples with h/d = 1.15 than for samples with h/d = 2.0. This trend was true in White et al.

(2005) study when 2 x 2 specimens were compared with 2.8 in. x 5.6 in. specimens, but not

when compared with Proctor size specimens.

In this study, the uncorrected results are reported for the UCS values.

Figure 3. Specimen failure after measurement of UCS

The cured specimens were tested in unsaturated and saturated condition. The specimens were

saturated using the vacuum saturated method as described in ASTM C593-06 (ASTM 2011a).

The specimens were placed on a perforated Plexiglas plate in a vacuum vessel (Figure 4), and the

chamber was evacuated using 24 in. of mercury for 30 minutes. Then the vacuum vessel was

flooded to a depth sufficient to cover the soil specimens. After one hour of soaking, the

specimens were removed from the vessel to conduct UCS testing. For samples that become

fragile and cannot be removed from water for UCS testing, the UCS is reported as 0 psi.

The cured specimens were tested in unsaturated and saturated condition. The specimens were

saturated using the vacuum saturated method as described in ASTM C593-06 (ASTM 2011a).

The specimens were placed on a perforated Plexiglas plate in a vacuum vessel (Figure 4), and the

chamber was evacuated using 24 in. of mercury for 30 minutes. Then the vacuum vessel was

flooded to a depth sufficient to cover the soil specimens. After one hour of soaking, the

specimens were removed from the vessel to conduct UCS testing. For samples that become

fragile and cannot be removed from water for UCS testing, the UCS is reported as 0 psi.

7

Figure 4. Vacuum saturation of cement stabilized specimens

Statistical Analysis

Multi-variate statistical analysis was conducted on the data generated from this study to assess

the influence of various independent parameters (i.e., soil index properties, cement content) in

predicting UCS, fines content passing the No. 200 sieve (F200), PI, and AASHTO group index.

Based on the analysis, prediction models were developed and presented in this report.

Analysis was performed using JMP statistical analysis software. The analysis was performed by

incorporating the independent parameters in a linear multiple regression model. The statistical

significance of the parameters were assessed using the t- and p-values associated with each

parameter. The selected criteria for identifying the significance of a parameter included p-value

≤ 0.05 is significant, ≤ 0.10 is possibly significant, > 0.10 = not significant, and t-value <-2 or

> +2 = significant. Higher the t- and p- values, greater is the statistical significance of the

parameter.

8

CHAPTER 3: MATERIALS

This chapter presents the soil index properties of the 28 embankment materials collected from 9

field projects in this study. The field project materials were obtained from Polk, Warren, Linn,

Pottawattamie, Mills, Woodbury and Scott Counties in Iowa. Embankment materials were

obtained from multiple test beds at each project sites. Gradation, Atterberg limits, specific

gravity, and compaction properties were tested for each material.

Table 3 to Table 8 provide a summary of the parent materials, particle size analysis, Atterberg

limits, specific gravity, soil classification, and Proctor compaction test results. The grain size

distribution curves of the embankment materials are separated by each project and shown in

Figure 5 to Figure 13. The embankment materials consisted of cohesive soils with glacial till at

three project sites and with western Iowa loess at four project sites. On one project site, granular

material consisted of alluvial sand from the Missouri river flood plain. Of the 25 cohesive

materials collected, 6 classified as select, 18 classified as suitable, and one classified as

unsuitable per Iowa DOT material suitability specification (Iowa DOT 2014). The three granular

soils collected were classified as suitable.

For cement stabilization, Type I/II portland cement was used in this study.

9

Table 3. Soil index properties of embankment materials obtained from Polk County

Parameter

Polk County

TB1

Polk County

TB2

Polk County

TB3

Polk County

TB4

5/29/2014 6/7/2014 8/5/2014 8/19/2014

Parent Material Glacial till Glacial till Glacial till Glacial till

Gravel content (%)

(> 4.75 mm) 0.4 3.9 2.6 1.8

Sand content (%)

(4.75 mm – 75 µm) 11.6 25.8 28.7 24.6

Silt content (%)

(75 µm – 2 µm) 66.4 34.7 45.8 50.9

Clay content (%) (<

2 µm) 21.6 35.6 22.9 22.7

LL (%) 49 45 36 34

PL (%) 28 34 20 17

PI (%) 21 11 16 17

AASHTO

classification A-7-6(21) A-7-5(8) A-6(9) A-6(11)

USCS classification CL CL CL CL

USCS Description Lean Clay Lean clay with

sand Sandy lean clay

Lean clay with

sand

Iowa DOT Material

Classification Suitable Suitable Suitable Suitable

Soil Color Olive Brown Olive Brown Very dark

greyish brown Olive Brown

Specific Gravity,

Gs 2.673 2.679 2.670 2.672

Std. Proctor, wopt

(%) 19.6 20.0 16.0 16.0

Std. Proctor, γdmax

(pcf) 103.9 104.0 110.0 110.6

Mod. Proctor, wopt

(%) 16.0 13.6 11.5 11.5

Mod. Proctor, γdmax

(pcf) 112.3 120.0 122.0 123.0

10

Table 4. Soil index properties of embankment materials obtained from Warren County and

Linn County 79

Parameter

Warren

County

TB1

Warren

County

TB2

Warren

County TB3

(Grey)

Warren

County TB3

(Brown)

Linn

County 79

6/3/2014 7/22/2014 8/4/2014 8/4/2014 6/6/2014

Parent Material Glacial till Glacial till Glacial till Glacial till weathered

loess

Gravel content (%)

(> 4.75 mm) 2.0 5.0 0.7 0.6 0.7

Sand content (%)

(4.75 mm – 75 µm) 27.5 31.6 18.7 29.2 46.0

Silt content (%)

(75 µm – 2 µm) 37.3 31.9 39.1 33.7 26.4

Clay content (%) (<

2 µm) 33.2 31.5 41.5 36.5 26.9

LL (%) 44 40 54 40 31

PL (%) 31 19 20 20 25

PI (%) 13 21 34 20 6

AASHTO

classification A-7-5(9) A-6(11) A-7-6(28) A-6(13) A-4(1)

USCS classification CL CL CH CL CL-ML

USCS Description Lean clay

with sand

Sandy lean

clay

Fat clay with

sand

Sandy lean

clay

Sandy silty

clay

Iowa DOT Material

Classification Suitable Select Unsuitable Suitable Suitable

Soil Color Olive

Brown

Light olive

Brown

Very dark

grey Olive Brown

Olive

Brown

Specific Gravity, Gs 2.676 2.673 2.715 2.674 2.684

Std. Proctor, wopt

(%) 16.5 15.0 21.0 17.0 13.5


(pcf) 111.1 113.8 102.0 109.5 117.4

Mod. Proctor, wopt

(%) 11.0 9.8 13.6 10.5 9.0


(pcf) 123.9 128.5 115.5 125.0 130.8

11

Table 5. Soil index properties of embankment materials obtained from Linn County 77

Parameter

Linn

County 77

TB1

Linn

County 77

TB2

Linn

County 77

TB3

Linn

County 77

TB4

Linn

County 77

TB5

6/6/2014 7/8/2014 7/15/2014 8/1/2014 9/8/2014

Parent Material Glacial till Glacial till Glacial till Glacial till Glacial till

Gravel content

(%) (> 4.75 mm) 1.8 1.3 11.3 1.1 2.0

Sand content (%)

(4.75 mm –

75 µm)

37.6 42.6 36.1 39.9 40.3

Silt content (%)

(75 µm – 2 µm) 32.9 30.9 31.2 35.6 34.8

Clay content (%)

(< 2 µm) 27.7 25.2 21.4 23.4 22.9

LL (%) 31 34 33 32 30

PL (%) 12 16 11 16 16

PI (%) 19 18 22 16 14

AASHTO

classification A-6(8) A-6(7) A-6(7) A-6(6) A-6(5)

USCS

classification CL CL CL CL CL

USCS

Description

Sandy lean

clay

Sandy lean

clay

Sandy lean

clay

Sandy lean

clay

Sandy lean

clay

Iowa DOT

Material

Classification

Select Select Select Select Select

Soil Color Very dark

grey Olive Brown

Very dark

grey

Very dark

grey

Very dark

grey

Specific Gravity,

Gs 2.683 2.670 2.673 2.672 2.674

Std. Proctor, wopt

(%) 12.9 13.0 12.0 11.7 12.6

Std. Proctor,

γdmax (pcf) 118.4 116.0 119.5 119.5 119.0

Mod. Proctor,

wopt (%) 8.8 9.0 8.0 8.1 8.6

Mod. Proctor,

γdmax (pcf) 130.8 129.5 131.0 132.1 130.0

12

Table 6. Soil index properties of embankment materials obtained from Pottawattamie

County and Woodbury County I-29

Parameter

Pottawattamie

County TB1

Pottawattamie

County TB2

Woodbury

County I-

29 TB1

Woodbury

County I-29

TB2

Woodbury

County I-29

TB3

7/2/2014 7/10/2014 7/9/2014 7/10/2014 8/7/2014

Parent Material Manufactured

materials

Manufactured

materials Alluvium Alluvium Alluvium

Gravel content

(%) (>

4.75 mm)

7.3 5.3 0.2 0.0 1.7

Sand content

(%) (4.75 mm

– 75 µm)

10.1 25.5 78.4 83.2 81.1

Silt content

(%) (75 µm –

2 µm)

56.2 48.0 15.5 12.6 11.6

Clay content

(%) (< 2 µm) 26.4 21.2 5.9 4.2 5.6

LL (%) 43 42 NP NP NP

PL (%) 18 19 NP NP NP

PI (%) 25 23 NP NP NP

AASHTO

classification A-7-6(20) A-7-6(14) A-2-4 A-2-4 A-2-4

USCS

classification CL CL SM SM SM

USCS

Description

Lean clay with

sand

Sandy lean

clay Silty sand Silty sand Silty sand

Iowa DOT

Material

Classification

Suitable Suitable Suitable Suitable Suitable

Soil Color Dark brown Very dark

greyish brown

Olive

Brown

Very dark

greyish

brown

Very dark

greyish

brown

Specific

Gravity, Gs 2.697 2.709 2.657 2.654 2.654

Std. Proctor,

wopt (%) 17.5 17.5 17.5 15.5 15.0

Std. Proctor,

γdmax (pcf) 106.0 106.3 102.5 102.8 104.5

Mod. Proctor,

wopt (%) 13.5 12.8 15.5 14.5 13.0

Mod. Proctor,

γdmax (pcf) 117.5 117.5 109.2 105.0 110.0

13

Table 7. Soil index properties of embankment materials obtained from Scott County and

Mills County

Parameter

Scott

County

TB1

Scott County

TB2

Scott County

TB3

Mills

County

TB1

Mills

County

TB2

7/16/2014 7/31/2014 9/19/2014 6/26/2014 6/26/2014

Parent Material Loess Loess Loess Loess Loess

Gravel content (%)

(> 4.75 mm) 0.1 1.0 2.0 0.1 3.9

Sand content (%)

(4.75 mm – 75 µm) 1.0 24.3 29.2 3.1 6.4

Silt content (%)

(75 µm – 2 µm) 72.9 45.5 45.9 70.6 34.9

Clay content (%) (<

2 µm) 26.0 29.2 22.9 26.2 54.8

LL (%) 39 35 28 38 36

PL (%) 32 24 17 34 31

PI (%) 7 11 11 4 5

AASHTO

classification A-4(10) A-6(8) A-6(5) A-4(7) A-4(6)

USCS classification CL-ML CL CL CL-ML CL-ML

USCS Description Silty Clay Lean clay

with sand

Sandy lean

clay Silty clay Silty clay

Iowa DOT Material

Classification Suitable Suitable Suitable Suitable Suitable

Soil Color Dark olive

brown

Dark

yellowish

brown

Olive Brown

Dark

yellow

brown

Brown

Specific Gravity,

Gs 2.680 2.672 2.673 2.725 2.726

Std. Proctor, wopt

(%) 16.5 15.5 13.0 17.0 16.0


(pcf) 108.0 111.1 119.5 108.5 110.8

Mod. Proctor, wopt

(%) 13.0 11.2 9.2 13.0 12.0


(pcf) 118.0 122.5 131.0 117.2 119.5

14

Table 8. Soil index properties of embankment materials obtained from Woodbury County

US 20

Parameter

Woodbury

County

(US20) TB1

Woodbury

County (US20)

TB2

Woodbury

County (US20)

TB3

Woodbury

County (US20)

TB4

9/26/2014 9/26/2014 10/18/2014 10/18/2014

Parent Material very deep loess very deep loess very deep loess very deep loess

Gravel content (%)

(> 4.75 mm) 0.0 0.0 0.1 0.0

Sand content (%)

(4.75 mm – 75 µm) 8.8 1.3 4.2 6.4

Silt content (%)

(75 µm – 2 µm) 68.8 73.3 69.6 72.0

Clay content (%) (<

2 µm) 22.4 25.4 26.1 21.6

LL (%) 32 35 35 31

PL (%) 25 27 23 24

PI (%) 7 8 12 7

AASHTO

classification A-4(7) A-4(9) A-6(12) A-4(7)

USCS classification CL-ML CL CL CL-ML

USCS Description Silty clay Lean clay Lean clay Silty clay

Iowa DOT Material

Classification Suitable Suitable Suitable Suitable

Soil Color Olive Brown Olive Brown Olive Brown Olive Brown

Specific Gravity,

Gs 2.717 2.679 2.673 2.720

Std. Proctor, wopt

(%) 16.0 18.4 18.0 16.0


(pcf) 110.0 106.0 106.7 110.5

Mod. Proctor, wopt

(%) 12.4 14.0 14.0 13.0


(pcf) 120.0 117.0 117.5 119.6

15

Figure 5. Grain size distribution of embankment materials obtained from Polk County

Figure 6. Grain size distribution of embankment materials obtained from Warren County

Grain Size (mm)

0.0010.010.1110100

Perc

ent

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3

TB4

3 in

.

2 in

.

1 in

.

3/4

in

.

3/8

in

.

#4

#10

#20

#40

#100

#200

#60

Grain Size (mm)

0.0010.010.1110100

Pe

rce

nt

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3 (Grey)

TB3 (Brown)

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#10

#20

#40

#100

#200

#60

16

Figure 7. Grain size distribution of embankment materials obtained from Linn County 77

Figure 8. Grain size distribution of embankment materials obtained from Linn County 79

Grain Size (mm)0.0010.010.1110100

Perc

ent

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3

TB4

TB5

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#10

#20

#40

#100

#200

#60

Grain Size (mm)

0.0010.010.1110100

Perc

ent

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

3 in

.

2 in

.

1 in

.

3/4

in

.

3/8

in

.

#4

#10

#20

#40

#100

#200

#60

17

Figure 9. Grain size distribution of embankment materials obtained from Mills County

Figure 10. Grain size distribution of embankment materials obtained from Pottawattamie

County

Grain Size (mm)

0.0010.010.1110100

Pe

rcen

t F

iner

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

3 in

.

2 in

.

1 in

.

3/4

in

.

3/8

in

.

#4

#10

#20

#40

#100

#200

#60

Grain Size (mm)

0.0010.010.1110100

Perc

ent

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#10

#20

#40

#100

#200

#60

18


County I-29

Figure 12. Grain size distribution of embankment materials obtained from Scott County

Grain Size (mm)

0.0010.010.1110100

Pe

rce

nt

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#1

0

#2

0

#4

0

#1

00

#2

00

#6

0

Grain Size (mm)

0.0010.010.1110100

Pe

rce

nt

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#10

#20

#40

#100

#200

#60

19


County US 20

Grain Size (mm)

0.0010.010.1110100

Perc

ent

Fin

er

(%)

0

10

20

30

40

50

60

70

80

90

100

TB1

TB2

TB3

TB4

3 in.

2 in.

1 in.

3/4

in.

3/8

in.

#4

#1

0

#2

0

#4

0

#1

00

#2

00

#6

0

20

CHAPTER 4: RESULTS AND DISCUSSION

This chapter presents the results obtained from laboratory tests and statistical analysis. A

summary of the F200, Atterberg limits, GI, and Iowa DOT material suitability classification

results for materials stabilized with different cement contents are presented in Table 9. Detailed

results are provided in Appendix A. In the following sections of this chapter, the results and

analysis are separately for F200, Atterberg limits, GI, and UCS, to present the influence of cement

stabilization on these properties.

Table 9. Summary of soil index properties and Iowa DOT suitability classifications at

different cement contents

County and

Test Bed

Cement

content (%) F200 (%) LL (%) PI (%) GI

Iowa DOT

Suitability

Polk TB1

0 88 49 21 21 suitable

4 74.1 41 13 10 suitable

8 64.5 40 8 5 suitable

12 53.1 40 0 0 suitable

Polk TB2

0 70.3 45 11 8 suitable

4 59.3 43 13 7 suitable

8 47.9 41 10 3 suitable

12 45.7 38 0 0 suitable

Polk TB3

0 68.7 36 16 9 suitable

4 58.5 34 6 2 suitable

8 41.1 35 0 0 suitable

12 32.3 36 0 0 suitable

Polk TB4

0 73.6 34 17 11 suitable

4 61.9 36 0 0 suitable

8 40.6 38 0 0 suitable

12 40.4 34 0 0 suitable

Warren TB1

0 70.5 44 13 9 suitable

4 60.4 38 14 7 suitable

8 36.8 41 0 0 suitable

12 27.4 38 0 0 suitable

Warren TB2

0 63.4 40 21 11 select

4 55.7 39 15 6 select

8 34.4 38 0 0 suitable

12 25.7 34 0 0 suitable

Warren TB3

0 80.6 54 34 28 unsuitable

4 70.7 42 17 11 suitable

8 51.8 44 12 4 suitable

12 31 40 0 0 suitable

21

County and

Test Bed

Cement

content (%) F200 (%) LL (%) PI (%) GI

Iowa DOT

Suitability

Linn 79 TB1

0 53.3 31 6 1 suitable

4 40.8 29 12 1 suitable

8 28.6 28 0 0 suitable

12 21.2 29 0 0 suitable

Linn 77 TB1

0 60.6 31 19 8 select

4 49.9 34 16 5 select

8 38.8 33 10 1 suitable

12 29.4 33 0 0 suitable

Linn 77 TB2

0 56.1 34 18 7 select

4 51.3 34 12 3 select

8 41 32 0 0 suitable

12 22.4 31 0 0 suitable

Linn 77 TB3

0 52.6 33 22 7 select

4 43.1 32 11 2 select

8 20.4 32 0 0 suitable

12 15.8 35 0 0 suitable

Linn 77 TB4

0 59 32 16 6 select

4 48 43 16 5 select

8 37 43 14 1 select

12 33.6 39 0 0 suitable

Linn 77 TB5

0 57.7 30 14 5 select

4 52.9 34 15 5 select

8 31.2 33 9 0 suitable

12 23.4 33 0 0 suitable

Pottawattamie

TB1

0 82.6 43 25 20 suitable

4 78.6 39 9 8 suitable

8 52.3 40 7 2 suitable

12 37.5 36 0 0 suitable

Pottawattamie

TB2

0 69.2 42 23 14 suitable

4 60.5 36 5 2 suitable

8 42.5 36 4 0 suitable

12 35.3 37 0 0 suitable

Mills TB1

0 96.8 38 4 7 suitable


8 49.8 34 2 0 suitable

12 34.5 36 0 0 suitable

Mills TB2

0 89.7 36 5 6 suitable

4 72.6 34 5 4 suitable

8 48.3 34 2 0 suitable

12 29.4 35 0 0 suitable

22

County and

Test Bed

Cement

content (%) F200 (%) LL (%) PI (%) GI

Iowa DOT

Suitability

Scott TB1

0 98.9 39 7 10 suitable

4 85.2 34 8 7 suitable

8 52.1 34 3 0 suitable

12 34.9 35 0 0 suitable

Scott TB2

0 74.7 35 11 8 suitable


8 46.9 32 0 0 suitable

12 40 34 0 0 suitable

Scott TB3

0 68.8 28 11 5 suitable

4 56.4 31 9 3 suitable

8 37.9 31 1 0 suitable

12 25.1 33 0 0 suitable

Woodbury

(US20) TB1

0 91.2 32 7 7 suitable

4 65.4 33 7 4 suitable

8 53.9 33 2 0 suitable

12 39 34 0 0 suitable

Woodbury

(US20) TB2

0 98.7 35 8 9 suitable

4 76.3 41 10 8 suitable

8 50.5 40 5 1 suitable

12 33.8 43 0 0 suitable

Woodbury

(US20) TB3

0 95.7 35 12 12 suitable

4 69.8 40 9 6 suitable

8 43.2 40 6 1 suitable

12 32.4 41 0 0 suitable

Woodbury

(US20) TB4

0 93.6 31 7 7 suitable

4 79.1 32 6 4 suitable

8 51.6 32 1 0 suitable

12 32.9 33 0 0 suitable

Woodbury

(I29) TB1

0 21.4 NV 0 0 suitable


8 9 NV 0 0 suitable

12 8.6 NV 0 0 select

Woodbury

(I29) TB2





Woodbury

(I29) TB3




12 8.3 NV 0 0 select

23

Fines Content (F200)

Results of F200 versus cement content are presented in Figure 14 and Figure 15. The results

indicated that F200 decreased with increasing cement content. Statistical analysis was conducted

to predict F200 after treatment as a function of cement content, F200 before treatment, and

Atterberg limits. Results are summarized in Table 10. Cement content, F200 before treatment,

and LL were found to be statistically significant. PI and PL parameters were not statistically

significant. Measured versus predicted F200 (after treatment) results from the multi-variate model

are presented in Figure 16. The model showed an R2 of about 0.9 and RMSE of about 7%.

Figure 14. F200 versus cement content

Cement content (%)

0 4 8 12

F2

00

(%)

0

20

40

60

80

100

Polk TB1

Polk TB2

Polk TB3

Polk TB4

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Warren TB1

Warren TB2

Warren TB3

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Linn 77 TB1

Linn 77 TB2

Linn 77 TB3

Linn 77 TB4

Linn 77 TB5

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Linn 79 TB1

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Pottawattamie TB1

Pottawattamie TB2

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Mills TB1

Mills TB2

24

Figure 15. F200 versus cement content (continued)

Table 10. Multi-variate analysis results to predict F200 after cement stabilization

Parameter Value t Ratio Prob> |t| R2

RMSE

Intercept 18.92 3.96 < 0.0001

0.898 6.588

Cement Content

(%) -3.74 -24.88 < 0.0001

F200 before

treatment (%) 0.607 13.23 < 0.0001

LL (%) 0.306 2.79 0.0064

Prediction

expression

F200 after treatment (%) = 18.92 - 3.74 x cement content (%) + 0.607 x F200 (%)

+ 0.306 x LL (%)

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Scott TB1

Scott TB2

Scott TB3

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Woodbury US20 TB1

Woodbury US20 TB2

Woodbury US20 TB3

Woodbury US20 TB4

Cement content (%)

0 4 8 12

F2

00

(%

)

0

20

40

60

80

100

Woodbury I-29 TB1

Woodbury I-29 TB2

Woodbury I-29 TB3

25

Figure 16. Comparison of measured F200 and predicted F200

Atterberg Limits

Plasticity charts showing relationship between LL and PI for unstabilized and stabilized soils

with 4%, 8%, and 12% cement content are shown in Figure 16 to Figure 20, respectively. F200

versus of PI results are shown in Figure 21. LL and PI versus cement content are presented in

Figure 22 to Figure 24.

With the exception of a few materials (Polk TB4, Linn 79, Linn 77 TB4), the LL and PI of all

materials decreased with increasing cement content. The one untreated soil classified as

“unsuitable”, classified as “suitable” after stabilized with cement. Some of the “select” untreated

soils classified as “suitable” after stabilized with cement, because of reduction in PI. All of the

soils classified as “suitable” at 12% cement content because of no plasticity.

Statistical analysis was conducted to predict PI after treatment as a function of cement content,

cement content, clay content, silt content, and LL. Results are summarized in Table 11. Cement

content and clay content were found to be statistically significant, while the remaining

parameters were not statistically significant. Measured versus predicted PI (after treatment)

results from the multi-variate model are presented in Figure 25. The model showed an R2 of

about 0.5 and RMSE of about 5%.

Measured F200 (%)

0 20 40 60 80 100 120

Pre

dic

ted

F2

00

(%

)

0

20

40

60

80

100

120F

200 after treatment(%) = 18.92-3.74 x PC(%)

+0.607 x F200

(%)+0.306 x LL(%)

R2=0.898 RMSE=6.588

1:1

26

Figure 17. Plasticity chart with results of unstabilized soils

Figure 18. Plasticity chart with results of 4% cement stabilized soils

Untreated Cohesive Soils

Liquid Limit, LL

0 10 20 30 40 50 60 70 80 90 100

Pla

sticity I

ndex,

PI

0

10

20

30

40

50

60

70

Select (6)

Suitable (17)

Unsuitable (1)

A-2-6

A-6

A-2-4

A-4

A-7-6

A-7-5

A-2-7

A-2-5

A-5

PI = L

L-30

4% PC Treated Cohesive Soils

Liquid Limit, LL

0 10 20 30 40 50 60 70 80 90 100

Pla

sticity I

ndex,

PI

0

10

20

30

40

50

60

70

Select (6)

Suitable (18)

A-2-6

A-6

A-2-4

A-4

A-7-6

A-7-5

A-2-7

A-2-5

A-5

PI = L

L-30

Note: PI = 0 represent Non-Plastic Soil

27




Liquid Limit, LL

0 10 20 30 40 50 60 70 80 90 100

Pla

sticity I

nd

ex,

PI

0

10

20

30

40

50

60

70

Select (1)

Suitable (23)

A-2-6

A-6

A-2-4

A-4

A-7-6

A-7-5

A-2-7

A-2-5

A-5

PI = L

L-30



Liquid Limit, LL

0 10 20 30 40 50 60 70 80 90 100

Pla

sticity I

ndex,

PI

0

10

20

30

40

50

60

70

Suitable (24)

A-2-6

A-6

A-2-4

A-4

A-7-6

A-7-5

A-2-7

A-2-5

A-5

PI = L

L-30


28

Figure 21. PI versus F200 for unstabilized and stabilized soils

F200

(%)

0 10 20 30 40 50 60 70 80 90 100

Pla

sticity I

nd

ex,

PI

0

5

10

15

20

25

30

35

40

Untreated

4% PC

8% PC

12% PC

0% PC

29

Figure 22. LL and PI versus cement content

30

Figure 23. LL and PI versus cement content (continued)

31

Figure 24. LL and PI versus cement content (continued-2)

Table 11. Multi-variate analysis results to predict PI after cement stabilization


RMSE

Intercept 8.664 5.85 < 0.0001

0.509 5.101 Cement Content

(%) -1.102 -10.04 < 0.0001

Clay content (%) 0.172 3.49 0.0007

Prediction

expression

F200 after treatment (%) = 8.664 – 1.102 x cement content (%) + 0.172 x Clay

content (%)

Note: Silt content, sand content, and LL were not statistically significant

32

Figure 25. Comparison of measured PI and predicted PI

AASHTO Group Index

GI versus cement content results are presented Figure 26 to Figure 27. For a majority of the soils,

the GI values decreased with increasing cement content. Statistical analysis was conducted to

predict GI after treatment as a function of cement content, clay content, silt content, F200, LL, and

PI. Results are summarized in Table 12. Cement content, F200, LL, and PI were found to be

statistically significant, while the remaining parameters were not statistically significant.

Measured versus predicted GI (after treatment) results from the multi-variate model are

presented in Figure 28. The model showed an R2 of about 0.7 and RMSE of about 3.

Measured PI (%)

0 10 20 30 40

Pre

dic

ted

PI

(%)

0

10

20

30

40PI(%)=8.664-1.102xPC(%)+0.172xClay(%)

R2=0.509 RMSE=5.1011:

1

33

Figure 26. AASHTO group index versus cement content

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

5

10

15

20

25

30

Warren TB1

Warren TB2

Warren TB3

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

5

10

15

20

25

Polk TB1

Polk TB2

Polk TB3

Polk TB4

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Linn 79 TB1

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

2

4

6

8

10

Linn 77 TB1

Linn 77 TB2

Linn 77 TB3

Linn 77 TB4

Linn 77 TB5

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

5

10

15

20

25

Pottawattamie TB1

Pottawattamie TB2

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

2

4

6

8

10

Mills TB1

Mills TB2

34

Figure 27. AASHTO group index versus cement content (continued)

Table 12. Multi-variate analysis results to predict GI after cement stabilization


RMSE

Intercept -4.540 -2.23 0.0281

0.708 2.774

Cement Content

(%) -0.844 -13.33 <0.0001

F200 (%) 0.069 2.85 0.0055

LL (%) 0.157 2.98 0.0164

PI (%) 0.172 2.45 0.0037

Prediction

expression

GI = - 4.540 – 0.844 x cement content (%) + 0.069 x F200 (%) + 0.157 x LL

(%) + 0.172 x PI (%)

Note: Silt content and clay content were not statistically significant.

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

2

4

6

8

10

12

Scott TB1

Scott TB2

Scott TB3

Cement content (%)

0 4 8 12

Gro

up

in

de

x

0

2

4

6

8

10

12

14

Woodbury US20 TB1

Woodbury US20 TB2

Woodbury US20 TB3

Woodbury US20 TB4

35

Figure 28. Comparison of measured group index and predicted group index

Unconfined Compressive Strength

Figure 29 to Figure 31 present the results of unsaturated and vacuum saturated UCS of the

materials at different cement contents. A linear regression line is fit to the data to define the

relationship between UCS and cement content. Results indicated increasing UCS with increasing

cement content, as expected. For a majority of the unstabilized materials, the soil specimens

became fragile after vacuum saturation and could not be retrieved from the vessel. For those

soils, UCS of 0 psi is reported herein. Vacuum saturated stabilized specimens resulted in UCS

measurements that were on average about 1.5 times lower than the unsaturated specimens. The

ratio of unsaturated and vacuum saturated UCS of stabilized specimens ranged from about 1.1 to

2.5.

Statistical analysis was conducted to predict unsaturated and vacuum saturated UCS as a

function of cement content, sand content, clay content, silt content, F200, LL, and PI. Results are

summarized in Table 13 and Table 14. Cement content, sand content, F200, and LL were found to

be statistically significant, while the remaining parameters were not statistically significant.

Measured versus predicted UCS results from the multi-variate model are presented in Figure 32

and Figure 33. The models showed an R2 of about 0.85 and RMSE of about 75 psi for vacuum

saturated UCS and 97 psi for unsaturated UCS.

Measured GI

0 5 10 15 20 25 30

Pre

dic

ted G

I

0

5

10

15

20

25

30GI=-4.540-0.844xPC(%)+0.069xF

200(%)

+0.157xLL(%)+0.172xPI(%)

R2=0.708 RMSE=2.7741:

1

36

Figure 29. Unsaturated and vacuum saturated UCS versus cement content

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Polk TB1

Linear (Polk TB1)

Polk TB2

Linear (Polk TB2)

Polk TB3

Linear (Polk TB3)

Polk TB4

Linear (Polk TB4)

Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Polk TB1

Linear (Polk TB1)

Polk TB2

Linear (Polk TB2)

Polk TB3

Linear (Polk TB3)

Polk TB4

Linear (Polk TB4)

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Warren TB1

Linear (Warren TB1)

Warren TB2

Linear (Warren TB2)

Warren TB3

LLinear (Warren TB3)

Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Warren TB1

Linear (Warren TB1)

Warren TB2

Linear (Warren TB2)

Warren TB3

Linear (Warren TB3)

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Linn 79 TB1

Linear (Linn 79 TB1)

Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Linn 79 TB1


37

Figure 30. Unsaturated and vacuum saturated UCS versus cement content (continued)

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Linn 77 TB1


Linn 77 TB2


Linn 77 TB3


Linn 77 TB4


Linn 77 TB5


Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000Linn 77 TB1


Linn 77 TB2


Linn 77 TB3


Linn 77 TB4


Linn 77 TB5


Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Pottawattamie TB1

Linear (Pottawattamie TB1)

Pottawattamie TB2


Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Pottawattamie TB1


Pottawattamie TB2


Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

1200

Mills TB1

Linear (Mills TB1)

Mills TB2

Linear (Mills TB2)

Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Mills TB1

Linear (Mills TB1)

Mills TB2

Linear (Mills TB2)

38

Figure 31. Unsaturated and vacuum saturated UCS versus cement content (continued-2)

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

1200

Scott TB1

Linear (Scott TB1)

Scott TB2

Linear (Scott TB2)

Scott TB3

Linear (Scott TB3)

Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

1200

Scott TB1

Linear (Scott TB1)

Scott TB2

Linear (Scott TB2)

Scott TB3

Linear (Scott TB3)

Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Woodbury US20 TB1

Linear (Woodbury US20 TB1)

Woodbury US20 TB2


Woodbury US20 TB3


Woodbury US20 TB4


Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Woodbury US20 TB1


Woodbury US20 TB2


Woodbury US20 TB3


Woodbury US20 TB4


Cement content (%)

0 4 8 12

Unsa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Woodbury I-29 TB1

Linear (Woodbury I-29 TB1)

Woodbury I-29 TB2


Woodbury I-29 TB3


Cement content (%)

0 4 8 12

Va

cu

um

sa

tura

ted

UC

S (

psi)

0

200

400

600

800

1000

Woodbury I-29 TB1


Woodbury I-29 TB2


Woodbury I-29 TB3


39

Table 13. Multi-variate analysis results to predict unsaturated UCS


Root Mean

square error

Intercept 1465.38 3.61 0.0005

0.848 97.418

Cement content

(%) 48.69 21.90 <0.0001

Sand (%) -13.26 -3.13 0.0023

F200 (%) -9.24 -2.35 0.0209

LL (%) -11.28 -6.77 <0.0001

Prediction

expression

UCS (psi) = 1465.38 + 48.69 x cement content (%)- 13.26 x Sand (%) - 11.28

x LL (%) - 9.24 x F200 (%)


Table 14. Multi-variate analysis results to predict vacuum saturated UCS


Root Mean

square error

Intercept 1151.32 3.7 0.0004

0.850 74.704

Cement content

(%) 37.33

21.89 <0.0001

Sand (%) -11.40 -3.51 0.0007

F200 (%) -7.70 -2.56 0.0123

LL (%) -8.37 -6.55 <0.0001

Prediction

expression

UCS (psi) = 1151.323 + 37.329 x cement content (%) - 11.401 x Sand (%) -

8.372 x LL (%) - 7.703 x F200 (%)


40

Figure 32. Comparison of measured unsaturated UCS and predicted unsaturated UCS

Figure 33. Comparison of measured vacuum saturated UCS and predicted UCS

Unsaturated UCS

Measured UCS (psi)

0 200 400 600 800 1000

Pre

dic

ted

UC

S (

psi)

0

200

400

600

800

1000UCS(psi)=1465.378+48.689xPC(%)-13.260xSand(%) -11.277xLL(%)-9.239xF200(%)

R2=0.848 RMSE=97.418 1:1

Vacuum saturated UCS

Measured UCS (psi)

0 200 400 600 800 1000

Pre

dic

ted

UC

S (

psi)

0

200

400

600

800

1000UCS(psi)=1151.323+37.329xPC(%)-11.401xSand(%) -8.372xLL(%)-7.703xF200(%)

R2=0.850 RMSE=74.704 1:1

41

CHAPTER 5: SUMMARY OF KEY FINDINGS

Results of a laboratory study focused on cement stabilization of 28 soils obtained from 9 active

construction sites in Iowa are presented in this report. The materials consisted of glacial till,

western Iowa loess, and alluvium sand. Type I/II portland cement was used for stabilization of

these materials. 2 x 2 specimens of stabilized and unstabilized materials were prepared, cured,

and tested for UCS with and without vacuum saturation. F200, AASHTO group index (GI), and

Atterberg limits were tested before and after stabilization. The results were analyzed using multi-

variate statistical analysis to assess influence of the various soil index properties on post-

stabilization material properties. Key findings from the test results and analysis are as follows:

F200 of the material decreased with increasing cement content for a majority of the soils. The

percent cement content, F200 before treatment, and liquid limit were found to be statistically

significant in predicting the F200 after treatment. The multi-variate model showed an R2 of

about 0.9 and RMSE of about 7% in predicting the F200 after treatment.

With the exception of a few materials, the liquid limit and plasticity index of all materials

decreased with increasing cement content. The one untreated soil classified as “unsuitable”,

classified as “suitable” after stabilized with cement. Some of the untreated soils that were

classified as “select”, classified as “suitable” after stabilized with cement. The

classifications changed because of reduction in plasticity index. All soils classified as

“suitable” at 12% cement content because of no plasticity. The percent cement content and

clay content parameters were found to be statistically significant in predicting the plasticity

index of materials after stabilization. The multi-variate model showed an R2 of about 0.5

and RMSE of about 5%.

The GI values decreased with increasing cement content for a majority of the soils. The

percent cement content, F200, liquid limit, and plasticity index parameters were found to be

statistically significant in predicting the group index values after treatment. The multi-

variate model showed an R2 of about 0.7 and RMSE of about 3.

The UCS of specimens increased with increasing cement content, as expected. The average

saturated UCS of the unstabilized materials varied between 0 and 57 psi. The average

saturated UCS of stabilized materials varied between 44 and 287 psi at 4% cement content,

108 and 528 psi at t 8% cement content, and 162 and 709 psi at 12% cement content. The

draft laboratory testing and evaluation procedure for cement stabilization mix design

provided in Appendix B targets a 100 psi saturated unconfined compressive strength. The

UCS of the saturated specimens was on average 1.5 times lower than of the unsaturated

specimens.

The percent cement content, sand content, fines content, and liquid limit were found to be

statistically significant in predicting unsaturated and vacuum saturated UCS. The models

showed an R2 of about 0.85 and RMSE of about 75 psi for vacuum saturated specimens and

97 psi for unsaturated specimens.

43

REFERENCES

AASHTO. (2004). “Standard specification for classification of soils and soil-aggregate mixtures

for highway construction purposes.” AASHTO M145-91, Annual Book of AASHTO

Standards, Washington, D.C.

ASTM. (2007). “Standard test methods for compressive strength of molded soil-cement

cylinders.” ASTM D1633-00, Annual Book of ASTM Standards, West Conshohocken,

PA.

ASTM. (2011). “Standard specification for fly ash and other pozzolans for use with lime for soil

stabilization.” ASTM C593-06, Annual Book of ASTM Standards, West Conshohocken,

PA.

ASTM. (2011). “Standard test method for classification of soils for engineering purposes

(Unified Soil Classification System).” ASTM D2487-11, Annual Book of ASTM

Standards, West Conshohocken, PA.

ASTM. (2012). “Standard test method for laboratory compaction characteristics of soil using

standard effort.” ASTM D698-12, Annual Book of ASTM Standards, West

Conshohocken, PA.

ASTM. (2012). “Standard test method for laboratory compaction characteristics of soil using

modified effort.” ASTM D1557-12, Annual Book of ASTM Standards, West

Conshohocken, PA.

ASTM. (2014a). “Standard test method for particle-size analysis of soils.” ASTM D422-63,

Annual Book of ASTM Standards, West Conshohocken, PA.

ASTM. (2014b). “Standard test method for liquid limit, plastic limit, and plasticity index of

soils.” ASTM D4318-10, Annual book of ASTM standards, West Conshohocken, PA.

ASTM. (2014c). “Standard test method for specific gravity of soil solids by water pycnometer.”

ASTM D854-14, Annual Book of ASTM Standards, West Conshohocken, PA.

ASTM. (2014d). “Standard test method for laboratory determination of water (moisture) content

of soil and rock by mass.” ASTM D2216-10, Annual Book of ASTM Standards, West

Conshohocken, PA.

ASTM. (2014e). “Standard test methods for determining the amount of material finer than 75-um

(No. 200) sieve in soils by washing.” ASTM D1140-14, Annual Book of ASTM

Standards, West Conshohocken, PA.

O’Flaherty, C.A., Edgar, C.E., Davidson, D.T. (1963). “The Iowa State Compaction Apparatus:

A Small Sample Apparatus for Use in Obtaining Density and Strength Measurements of

Soils and Soil-Additives,” Special Report Prepared for Presentation at the 42nd

Annual

Meeting of the Highway Research Board, Contribution No. 63-5 from the Soil Research

Laboratory, Iowa Engineering Experiment Station, Iowa State University, Ames, Iowa.

Iowa DOT. (2014). “General supplemental specifications for highway and bridge construction.”

GS-12005, Section 2102, Ames, IA, pp. 17.

Li, H., and Sego, D.C. (1999). “Soil compaction parameters and their relationship with soil

physical properties.” 52nd Canadian Geotechnical Conference, Regina, SK, Canada, pp.

517-524.

White, D.J., Harrington, D., and Thomas, Z. (2005). “Fly ash soil stabilization for non-uniform

subgrade soils, Volume I: Engineering properties and construction guidelines.” Final

report for Iowa Highway Research Board Project TR-461. Center for Portland Cement

Concrete Pavement Technology (PCC Center), Ames, IA.

44

Winterkorn, H.F., and Pamukcu, S. (1990). “Soil stabilization and grouting.” Chapter 9 in

Foundation Engineering Handbook, ed. By H-Y Fang, 2nd Edition, Van Nostrand

Reinhold, New York.

45

APPENDIX A: LABORATORY TEST RESULTS

46

Test Bed

Treated soil properties Untreated soil properties

Cement

content

(%)

UCS (psi) Atterberg limits

F200

Group

index

Iowa DOT

Material

Suitability

Gravel

content

(%)

Sand

content

(%)

Silt

content

(%)

Clay

content

(%)

USCS

Classification

AASHTO

Classification Unsaturated Saturated LL PL PI

Polk TB1

0 50.4 8.5 49 28 21 88 21 suitable

0.4 11.6 66.4 21.6 CL A-7-6(21) 4 174.3 78.6 41 28 13 74.1 10 suitable

8 279.9 230.6 40 32 8 64.5 5 suitable

12 409.8 320.7 40 NP 0 53.1 0 suitable

Polk TB2

0 36.8 18.7 45 34 11 70.3 8 suitable

3.9 25.8 34.7 35.6 CL A-7-5(8) 4 120.2 54.3 43 30 13 59.3 7 suitable

8 324 187.1 41 31 10 47.9 3 suitable

12 442 265.2 38 NP 0 45.7 0 suitable

Polk TB3

0 9.6 56.9 36 20 16 68.7 9 suitable

2.6 28.7 45.8 22.9 CL A-6(9) 4 224.1 134.2 34 28 6 58.5 2 suitable

8 336.7 251.7 35 NP 0 41.1 0 suitable

12 519.4 351.2 36 NP 0 32.3 0 suitable

Polk TB4

0 54.2 8.3 34 17 17 73.6 11 suitable

1.8 24.6 50.9 22.7 CL A-6(11) 4 261.8 135.1 36 NP 0 61.9 0 suitable

8 438.5 313.6 38 NP 0 40.6 0 suitable

12 634.4 461.2 34 NP 0 40.4 0 suitable

Warren TB1

0 59.3 0 44 31 13 70.5 9 suitable

2 27.5 37.3 33.2 CL A-7-5(9) 4 181.9 107.9 38 24 14 60.4 7 suitable

8 431.1 228.6 41 NP 0 36.8 0 suitable

12 686.9 359.7 38 NP 0 27.4 0 suitable

Warren TB2

0 38.3 0 40 19 21 63.4 11 select

5 31.6 31.9 31.5 CL A-6(11) 4 223.3 103.7 39 24 15 55.7 6 select

8 413.7 213.3 38 NP 0 34.4 0 suitable

12 512 317.2 34 NP 0 25.7 0 suitable

Warren TB3

0 38.7 0 54 20 34 80.6 28 unsuitable

0.7 18.7 39.1 41.5 CH A-7-6(28) 4 150.8 68 42 25 17 70.7 11 suitable

8 201 147.8 44 32 12 51.8 4 suitable

12 305.6 239.7 40 NP 0 31 0 suitable

Linn 79 TB1

0 48.9 0 31 25 6 53.3 1 suitable

0.7 46 26.4 26.9 CL-ML A-4(1) 4 257.7 118.4 29 17 12 40.8 1 suitable

8 475.8 296.5 28 NP 0 28.6 0 suitable

12 492.2 408.8 29 NP 0 21.2 0 suitable

Linn 77 TB1

0 60 0 31 12 19 60.6 8 select

1.8 37.6 32.9 27.7 CL A-6(8) 4 224.2 114.7 34 18 16 49.9 5 select

8 397.1 255.5 33 23 10 38.8 1 suitable

12 414.3 325.6 33 NP 0 29.4 0 suitable

Linn 77 TB2

0 53.1 0 34 16 18 56.1 7 select

1.3 42.6 30.9 25.2 CL A-6(7) 4 233.2 121.5 34 22 12 51.3 3 select

8 466.6 290.4 32 NP 0 41 0 suitable

12 605.3 456.7 31 NP 0 22.4 0 suitable

Linn 77 TB3 0 67.5 0 33 11 22 52.6 7 select 11.3 36.1 31.2 21.4 CL A-6(7)

47

4 305.6 219.3 32 21 11 43.1 2 select

8 676.6 472.9 32 NP 0 20.4 0 suitable

12 863.4 598 35 NP 0 15.8 0 suitable

Linn 77 TB4

0 68.6 0 32 16 16 59 6 select

1.1 39.9 35.6 23.4 CL A-6(6) 4 146.8 78.8 43 27 16 48 5 select

8 281.9 163.1 43 29 14 37 1 select

12 436 271.9 39 NP 0 33.6 0 suitable

Linn 77 TB5

0 47.1 0 30 16 14 57.7 5 select

2 40.3 34.8 22.9 CL A-6(5) 4 264.4 105.2 34 19 15 52.9 5 select

8 424.2 269.6 33 24 9 31.2 0 suitable

12 635.5 355.8 33 NP 0 23.4 0 suitable

Pottawattamie

TB1

0 63.9 5.3 43 18 25 82.6 20 suitable

7.3 10.1 56.2 26.4 CL A-7-6(20) 4 260.7 160.6 39 30 9 78.6 8 suitable

8 447.6 324.9 40 33 7 52.3 2 suitable

12 654.6 486.8 36 NP 0 37.5 0 suitable

Pottawattamie

TB2

0 49.3 0 42 19 23 69.2 14 suitable

5.3 25.5 48 21.2 CL A-7-6(14) 4 208.4 155.5 36 31 5 60.5 2 suitable

8 287.2 255.8 36 32 4 42.5 0 suitable

12 296 211.9 37 NP 0 35.3 0 suitable

Mills TB1

0 53.9 0 38 34 4 96.8 7 suitable

0.1 3.1 70.6 26.2 CL-ML A-4(7) 4 268.8 224 35 27 8 88 8 suitable

8 762.9 528.1 34 32 2 49.8 0 suitable

12 903.1 709.1 36 NP 0 34.5 0 suitable

Mills TB2

0 55.4 1.7 36 31 5 89.7 6 suitable

3.9 6.4 34.9 54.8 CL-ML A-4(6) 4 337.1 286.9 34 29 5 72.6 4 suitable

8 632.4 464.3 34 32 2 48.3 0 suitable

12 747.7 624.8 35 NP 0 29.4 0 suitable

Scott TB1

0 59.2 5.8 39 32 7 98.9 10 suitable

0.1 1 72.9 26 CL-ML A-4(10) 4 257.3 167.7 34 26 8 85.2 7 suitable

8 533.2 353 34 31 3 52.1 0 suitable

12 686.7 519 35 NP 0 34.9 0 suitable

Scott TB2

0 44 6.6 35 24 11 74.7 8 suitable

1 24.3 45.5 29.2 CL A-6(8) 4 299.8 197.2 33 27 6 61 2 suitable

8 608.6 484.9 32 NP 0 46.9 0 suitable

12 820.7 605.9 34 NP 0 40 0 suitable

Scott TB3

0 48.4 5.8 28 17 11 68.8 5 suitable

2 29.2 45.9 22.9 CL A-6(5) 4 333 244.3 31 22 9 56.4 3 suitable

8 696.6 461.5 31 30 1 37.9 0 suitable

12 980.6 692.4 33 NP 0 25.1 0 suitable

Woodbury

(US20) TB1

0 60.2 9.3 32 25 7 91.2 7 suitable

0 8.8 68.8 22.4 CL-ML A-4(7) 4 292.3 184.4 33 26 7 65.4 4 suitable

8 525.8 429.3 33 31 2 53.9 0 suitable

12 789.7 554.3 34 NP 0 39 0 suitable

Woodbury

(US20) TB2

0 59.7 0 35 27 8 98.7 9 suitable 0 1.3 73.3 25.4 CL A-4(9)

4 278.6 189.4 41 31 10 76.3 8 suitable

48

8 488.4 341 40 35 5 50.5 1 suitable

12 663.3 484 43 NP 0 33.8 0 suitable

Woodbury

(US20) TB3

0 52.9 3.8 35 23 12 95.7 12 suitable

0.1 4.2 69.6 26.1 CL A-6(12) 4 288.7 169 40 31 9 69.8 6 suitable

8 534.4 343 40 34 6 43.2 1 suitable

12 735.7 513.7 41 NP 0 32.4 0 suitable

Woodbury

(US20) TB4

0 63.3 4.4 31 24 7 93.6 7 suitable

0 6.4 72 21.6 CL-ML A-4(7) 4 339.8 196 32 26 6 79.1 4 suitable

8 588.6 431.6 32 31 1 51.6 0 suitable

12 815 572.2 33 NP 0 32.9 0 suitable

Woodbury (I29)

TB1

0 0 0 NV NP NP 21.4 0 suitable

0.2 78.4 15.5 5.9 SM A-2-4 4 94.7 81.7 NV NP NP 9.3 0 suitable

8 268.6 234.9 NV NP NP 9 0 suitable

12 506.2 439.8 NV NP NP 8.6 0 select

Woodbury (I29)

TB2


0 83.2 12.6 4.2 SM A-2-4 4 54.6 43.8 NV NP NP 7.7 0 suitable

8 120.2 108.2 NV NP NP 7.1 0 suitable

12 187.4 161.7 NV NP NP 7.4 0 suitable

Woodbury (I29)

TB3


1.7 81.1 11.6 5.6 SM A-2-4 4 100 72.5 NV NP NP 8.2 0 suitable

8 238.3 211.4 NV NP NP 9.5 0 suitable

12 414.6 398.6 NV NP NP 8.3 0 select

49

APPENDIX B: IOWA DOT PROPOSED INSTRUCTIONAL MEMORANDUM FOR

CEMENT STABILIZATION OF SOILS

CEMENT STABILIZATION OF SOILS GENERAL This procedure describes procedures for sampling and testing, and requirements for submittal and approval of mix design for cement stabilized soils.

SAMPLING AND MATERIALS Each soil sample to be used in chemical stabilization shall be 75 pounds (35 kg). This sample size will also provide for tests to be performed according to Materials IM 545. The cement used for stabilization shall meet the requirements of Type I or I/II from Section 4101.

SAMPLE PREPARATION AND TESTING

Laboratory tests on untreated soil shall be performed according to Materials IM 545. The material suitability should be classified in accordance with Section 2102. Additionally, sulfate content of the soil shall be determined per AASHTO T290. If the soil consists of soluble sulfate content > 3,000 ppm or the material classifies as unsuitable, chemical stabilization shall not be performed unless consulted with the engineer. For each soil type, prepare three samples each for the following four mixes:

Mix 1: Untreated soil

Mix 2: 2% cement

Mix 3: 4% cement

Mix 4: 6% cement. To determine the quantity of cement to add to the soil, multiply the cement percentage by the dry weight of the soil. Use cement that is from the same source(s) that will be used during construction. First, the moisture-density relationship of the different mixtures shall be determined. Then, unconfined compressive strength testing shall be performed at target moisture contents, as described below. Moisture-Density Relationship The moisture versus dry density relationship of untreated and cement-treated samples shall be determined using one of the following alternatives:

50

Alternative 1:

Untreated Samples: The maximum dry density and optimum moisture content of the untreated samples shall be determined using standard Proctor test in accordance with ASTM D698-12 [Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Standard Effort (12,400 ft-lb/ft3 (600 kN-m/m3)). A minimum 3-point Proctor is recommended.

Treated Samples: The maximum dry density and optimum moisture content shall be determined in accordance with ASTM D558-11 [Standard Test Methods for Moisture-Density (Unit Weight) Relations of Soil-Cement Mixtures]. All treated samples must be compacted within 1 hour of mixing. A minimum 3-point Proctor is recommended.

Alternative 2: The maximum dry density and optimum moisture content of untreated and treated samples shall be determined using the Iowa State University 2” by 2” Moisture-Density Test Method, per O’Flaherty et al. (1963). In preparing samples using the 2” by 2” method, use the following table for guidance on the total number of drop-hammer blows depending on the soil type to obtain results similar to the standard Proctor test.

Total number of drop-hammer blows

Soil type (based on AASHTO system)

6 A7 and A6

7 A4

14 A3, A2, and A1

Alternative 3: First, determine the optimum moisture content of the untreated soil using standard Proctor test in accordance with ASTM D698-12 [Standard Test Method for Laboratory Compaction Characteristics of Soil Using Standard Effort (12400 ft-lbf/ft3 (600 kN-m/m3))]. Then use the following equation to determine the optimum moisture content of treated samples, by using a water to cement (w/c) ratio of 0.25: wopt soil + cement = [(% cement added by weight) x (w/c ratio)] + wopt soil Unconfined Compressive Strength The unconfined compressive strength (UCS) tests shall be conducted on compacted samples at respective optimum moisture contents for untreated and treated soils, in accordance with ASTM D1633-00 (2007) [Standard Test Methods for Compressive Strength of Molded Soil-Cement Cylinders]. As an alternative, tests can be conducted on 2” by 2” samples prepared per Alternative 2 above. For each mix, prepare three samples for UCS testing for a total of twelve samples. Wrap each sample immediately after compaction with a plastic wrap and aluminum foil and store in a moisture-proof and airtight bag. All treated samples shall be cured at 100oF (38oC) for 7 days. Untreated samples shall be cured for no more than 24 hours.

51

After curing, all samples shall be vacuum saturated in accordance with ASTM C593-06 (2011) Section 11 [Standard Specification for Fly Ash and Other Pozzolans for Use with Lime for Soil Stabilization]. For samples that become fragile and cannot be retrieved from water for UCS testing, report the UCS as 0 psi. Target cement content determination The data obtained from UCS testing shall be plotted on a graph with cement content on x-axis and saturated UCS on y-axis. The average UCS of three samples shall be reported on the y-axis. The cement content corresponding to a saturated UCS of 100 psi shall be determined. 0.5% cement shall be added to determine the target cement content for the field application, as illustrated in Figure 1. Measurement of the true UCS is considered for specimens with height to diameter (h/d) ratio of 2.0. The Proctor size specimens prepared in accordance with ASTM D1633 have h/d = 1.15 and 2 x 2 specimens have h/d = 1.0. The UCS determined from these two sample sizes must be corrected as shown in Figure 2, to calculate the true UCS for sample with h/d = 2.0. The correction factors are based on a laboratory study on fly ash soil mixtures by White et al. (2005).

Figure 1. Determination of target cement content for field application

Figure 2. Correction factors for UCS determined from different sample sizes

Cement content (%)

0 1 2 3 4 5 6 7 8

Sa

tura

ted

UC

S (

psi)

0

50

100

150

200

Target cement contentfor field

+0.5%

Target Saturated UCS = 100 psi

Std. Sample

A: h/d = 2

2x2 Sample

B: h/d = 1

Proctor Sample

C: h/d = 1.15

UCS: 100 psi 111 psi 95 psi

Relationship between UCS of standard sample size, 2x2

sample, and Proctor size sample are based on experimental

test results documented by White et al. (2005)

B = 1.11 x A

C = 0.86 x B

52

REPORTS Each report shall contain the following for untreated soil:

Sample ID number and location

Atterberg Limits

Percent Gravel, Sand, Silt, and Clay

Textural classification

AASHTO classification

Proctor density and optimum moisture

Percent Carbon Content

Sieve analysis (Percent Passing)

Sulfate content

Additionally, each report shall contain the following for untreated and treated soils (for each soil type, there will be a total of twelve samples):

Percent cement added in each mixture

Maximum dry density and optimum moisture content, and the alternative procedure followed as described in this IM.

Unconfined compressive strength – for each sample Submit a graph similar to Figure 1 with average saturated UCS versus % of cement in the mixture with the recommended rate of chemical stabilization for review and approval by the Engineer.

REFERENCES

White, D.J., Harrington, D., and Thomas, Z. (2005). “Fly ash soil stabilization for non-uniform subgrade soils, Volume I: Engineering properties and construction guidelines.” Final report for Iowa Highway Research Board Project TR-461. Center for Portland Cement Concrete Pavement Technology (PCC Center), Ames, IA.

O’Flaherty, C.A., Edgar, C.E., Davidson, D.T. (1963). “The Iowa State Compaction Apparatus: A Small Sample Apparatus for Use in Obtaining Density and Strength Measurements of Soils and Soil-Additives,” Special Report Prepared for Presentation at the 42nd Annual Meeting of the Highway Research Board, Contribution No. 63-5 from the Soil Research Laboratory, Iowa Engineering Experiment Station, Iowa State University, Ames, Iowa.

cement stabilization of embankment materials final report

Documents